Share
Export Citation
Adaptive Gain Based Threshold Optimization in Stochastic Hybrid Automatic Transfer Switching for Residential Energy Systems
Makkulau A.
International Review of Electrical Engineering
Q3Abstract
This study develops a numerical simulation framework for a stochastic hybrid model of an adaptive Automatic Transfer Switch (ATS) in residential photovoltaic-battery systems subject to uncertain grid availability. The proposed formulation integrates continuous battery State-of-Charge (SOC) dynamics, discrete switching logic, and probabilistic grid disturbances modeled as a Bernoulli process. Unlike conventional fixed SOC window strategies, the lower SOC boundary is parameterized through a gain-dependent adaptive function linked to estimated grid reliability. The resulting system is analyzed as a gain-tunable hybrid dynamical model with a switching surface influenced by stochastic disturbances. Discrete-time simulations have been implemented in Python to evaluate system behavior under stable and unstable grid scenarios. A structured parametric exploration has been conducted for k = {0.0, 0.2, 0.3, 0.4, 0.6} and system performance has been assessed by using a normalized multi-objective index combining Loss of Load Probability (LOLP), switching function frequency, and battery cycling approximation. Simulation results reveal quasi-convex cost behavior within the bounded gain domain, with an optimal region around k ≈ 0.2, which achieves the best trade-off, reducing LOLP from 132 to 123 under unstable grid conditions. These findings highlight the role of adaptive switching boundary modeling in shaping reliability-stress trade-offs within stochastic hybrid residential energy systems.